HAL Id: hal-02550004https://hal.univ-lorraine.fr/hal-02550004
Submitted on 27 Dec 2020
HAL is a multi-disciplinary open accessarchive for the deposit and dissemination of sci-entific research documents, whether they are pub-lished or not. The documents may come fromteaching and research institutions in France orabroad, or from public or private research centers.
L’archive ouverte pluridisciplinaire HAL, estdestinée au dépôt et à la diffusion de documentsscientifiques de niveau recherche, publiés ou non,émanant des établissements d’enseignement et derecherche français ou étrangers, des laboratoirespublics ou privés.
Distributed under a Creative Commons Attribution - NonCommercial - NoDerivatives| 4.0International License
Zn2+ leakage and photo-induced reactive oxidativespecies do not explain the full toxicity of ZnO core
Quantum DotsXavier Bellanger, Raphaël Schneider, Clément Dezanet, Boussad Arroua,
Lavinia Balan, Patrick Billard, Christophe Merlin
To cite this version:Xavier Bellanger, Raphaël Schneider, Clément Dezanet, Boussad Arroua, Lavinia Balan, et al..Zn2+ leakage and photo-induced reactive oxidative species do not explain the full toxicity ofZnO core Quantum Dots. Journal of Hazardous Materials, Elsevier, 2020, 396, pp.122616.�10.1016/j.jhazmat.2020.122616�. �hal-02550004�
1
Zn2+
leakage and photo-induced reactive oxidative species
do not explain the full toxicity of ZnO core Quantum Dots
Xavier Bellanger a, Raphaël Schneider
b*, Clément Dezanet
b, Boussad Arroua
a, Lavinia
Balan c, Patrick Billard
d, Christophe Merlin
a*
a Université de Lorraine, CNRS, LCPME, F-54000 Nancy, France.
b Université de Lorraine, CNRS, LRGP, F-54000 Nancy, France.
c Institut de Science des Matériaux de Mulhouse (IS2M), CNRS, UMR 7361, 15 rue Jean
Starcky, 68093 Mulhouse, France
d Université de Lorraine, CNRS, LIEC, F-54000 Nancy, France
* to whom correspondence should be addressed
KEYWORDS. Metal oxide nanoparticles; ZnO Quantum Dots; Nanotoxicity; Photo-induced
Reactive oxygen species; Cell surface damages.
ABSTRACT
Metal oxide nanoparticles (NPs), and among them metal oxides Quantum Dots (QDs),
exhibit a multifactorial toxicity combining metal leaching, oxidative stress and possibly
direct deleterious interactions, the relative contribution of each varying according to the NP
composition and surface chemistry. Their wide use in public and industrial domains requires
a good understanding and even a good control of their toxicity. To address this question, we
engineered ZnO QDs with different surface chemistries, expecting that they would exhibit
different photo-induced reactivities and possibly different levels of interaction with biological
materials. No photo-induced toxicity could be detected on whole bacterial cell toxicity
assays, indicating that ROS-dependent damages, albeit real, are hidden behind a stronger
source of toxicity, which was comforted by the fact that the different ZnO QDs displayed the
same level of cell toxicity. However, using in vitro DNA damage assays based on
quantitative PCR, significant photo-induced reactivity could be measured precisely, showing
that different NPs exhibiting similar inhibitory effects on whole bacteria could differ
2
dramatically in terms of ROS-generated damages on biomolecules. We propose that direct
interactions between NPs and bacterial cell surfaces prime over any kind of intracellular
damages to explain the ZnO QDs toxicity on whole bacterial cells.
1. INTRODUCTION
Late developments in nanotechnologies are experiencing such progresses that they led
to the engineering of a plethora of nanoparticles (NPs) with unique properties already
fulfilling various tasks in modern life as they massively enter in industrial processes and
consumer products [1,2]. Among them, metal oxide NPs, such as zinc oxide (ZnO) NPs, are
currently focusing considerable attentions due to their high potential for applications such as
catalysis, sensing and environmental remediation [3-6]. Nevertheless, the outstanding
properties associated to the nanoscale of these particles has led to the rapid development of
innovative technologies that surpasses the full understanding of the adverse effects of
nanomaterials when health and environmental impacts are considered. However, gaining a
wide public acceptance for such NPs necessarily implies a deep knowledge and a fine control
of their toxicity in order to limit the risks for human and environmental health.
Regarding metal oxide toxicity, a plethora of reports are regularly opposing the
production of reactive oxygen species (ROS), the release of metal cations and even direct
interaction between the NPs and the cell surface, a long standing debate that reflects the
complexity of the underlying toxicity mechanisms involved [7-15]. The toxicity of ZnO NPs
may be partially attributed to the dissolution of the nanocrystals, which depends on pH, on
the size of the particles and on the medium composition, and the associated release of Zn2+
ions may enter the cells and damage cellular components such as proteins and DNA [16-18].
Recent reports highlight also the key role played by the ROS generated by ZnO NPs under
light irradiation and even in the dark [8,9,19-23]. ZnO is a semiconductor with a bandgap of
3.37 eV in the bulk state and is able to be activated by wavelengths equal of below 368 nm.
The associated oxido-reduction reactions occurring with O2 and H2O at the surface of ZnO
NPs generate hydroxyl (•OH), superoxide (O2
•-) and hydroperoxyl (HO2
•) radicals and H2O2
that cause membrane damages by lipid peroxidation [24]. Moreover, these membrane
damages favor the internalization of ZnO NPs and lead to cell death. Noteworthy is also that
these photo-generated ROS not only oxidize chemical and biological species located in the
vicinity of the NPs but may also affect the ligands at the periphery of the NPs and cause NPs
aggregation and/or dissolution [25,26]. Although the photoxicity of ZnO NPs is directly
3
related to the level of ROS produced, it also depends on the proximity between the source of
ROS and the cells, and thus the level of interaction between the cells and the NP for which
the surface chemistry is a key parameter [27]. Clearly, future developments of metal oxide-
based nanotechnologies will require a better understanding of the different sources of toxicity
and their relative contribution, in order to engineer NPs of controlled and predictable toxicity.
In this study, we investigated the relationship between the surface chemistry of ZnO
core QDs and the toxicity of the nanoparticles against Escherichia coli using whole cell
assays. Although surprising, different ZnO core QDs exhibiting different levels of reactivity
did not differ significantly in terms of toxicity when assessed using whole cell luminescent
biosensors. This led us to develop a molecular-based assay where DNA degradation was used
as surrogate to evaluate the level of ROS-generated damages on biomolecules. This approach
allows pointing out different levels of deleterious effects that were otherwise undetectable
when using whole cell toxicity assays.
2. MATERIALS AND METHODS
2.1. Synthesis of ZnO core QDs and evaluation of their stability
3-Aminopropyltrimethoxysilane (APTMS), (3-glycidyloxypropyl)trimethoxysilane
(GLYMO) and the sulfobetaine zwitterionic (SBZ) ligand were used to disperse oleate-
capped ZnO QDs in water (Fig. 1). The synthetic protocols used to prepare ZnO@APTMS,
ZnO@GLYMO, ZnO@SBZ QDs are detailed in the Supporting Information. ZnO QDs were
characterized by transmission electron microscopy (TEM), X-ray diffraction (XRD), X-ray
photoelectron spectroscopy (XPS) and UV-visible and fluorescence spectroscopy.
Fig. 1. Structures of the APTMS, GLYMO and SBZ ligands used to disperse ZnO QDs in
water.
4
The stability of the ZnO QDs with respect to dissolution was assessed by measuring the
amount of leaked Zn2+
(ICP-OES) recovered from the dialysate of an 8–10 kD dialysis device
(Float-A-Lyzer G2; Spectra/Por), see Supporting information for details [14].
2.2. Evaluation of QDs-generated ROS
The amount of •OH, O2
•- radicals and of H2O2 produced by the QDs, with or without photo-
induction, were determined using disodium terephthalate (DST), nitroblue tetrazolium (NBT)
and crystal violet/horseradish peroxidase assays, respectively, as already reported by Moussa
et al. [23] (Supporting information). All experiments were conducted using aqueous
dispersions of QDs with an UV-visible absorption of 0.1 at 330 nm. Results were corrected
relative to the Zn content of each sample determined by inductively coupled plasma-optical
emission spectrometer (ICP-OES) analyses. ROS measurements were mostly carried out in
water and cannot be transposed directly into a possible toxicity level against bacteria in
culture media as production of ROS and their availability are influenced by the surrounding
biological medium. Nevertheless, comparative assays made using the ZnO@APTMS QDs
showed that photogenerated-ROS detected in MOPS culture medium were about 30% of the
value measured in water (Supporting information, section 2.4), which remains far from being
negligible.
2.3. Evaluation of ZnO QDs toxicity
Toxicity tests on live bacterial cells were based on bioluminescence extinction assays of the
E. coli strain MG1655(pUCD07) as described elsewhere [14] (see Supporting information).
Typically, the luminescent bacterium was cultivated in a MOPS mineral medium known to
limit metal chelation, and supplemented with gluconate as sole carbon source. Cultures were
exposed to a set of ZnO QDs, with or without exposure of a full spectrum neon lighting (18W
T8 “Solar Reptil Sun” from JBL, illuminance ca. 9 klux). The extinction of the bacterial
luminescence was recorded on plate reader and normalized to the biomass (OD600).
5
2.4. Evaluation of QD-generated stresses
Exposure to free Zn2+
, oxidative stress or DNA damaging stress, were assessed using whole
cell biosensor assays with dedicated E. coli strains. These consist of genetically modified
bacteria specifically engineered to emit a bioluminescent signal when the stress is perceived
by the live bacteria (transcriptional fusions of regulated promoters to lux reporter genes).
Strains construction and use are detailed in the Supporting information.
2.5 Assessing QDs reactivity against DNA
Photo-induced damage to DNA assays consisted in exposing 500 ng of a well-defined
pBELX plasmid DNA [28], for 10 min, to known concentrations of ZnO QDs, with or
without exposure to a full spectrum neon lighting as described for the toxicity test. The
exposed DNA was further purified using a QIAquick® PCR purification kit (Qiagen) and
used as template for the quantification of unaltered molecules by qPCR (Supporting
information). DNA treated similarly with free Zn2+
did not show significant damage nor
recovery limitation.
3. RESULTS AND DISCUSSION
3.1. ZnO QDs synthesis and characterization
Oleic acid-capped ZnO QDs were prepared via a sol-gel process according to our previous
reports [ref]. To disperse the hydrophobic NPs in aqueous solution, a ligand exchange was
conducted using 3-aminopropyltrimethoxysilane (APTMS), (3-
glycidyloxypropyl)trimethoxysilane (GLYMO) or with a sulfobetaine zwitterionic (SBZ)
ligand (Fig. 1) in order to engineer NPs presenting different level of photo-induced ROS
production, as well as different surface chemistries enabling to develop different types of
interactions with bacterial cells (Supporting Information).
The size and the shape of ZnO QDs were characterized by TEM (Fig. 2a-c). The micrographs
show that the dots were relatively monodispersed and the average size of the
spherical/ellipsoidal NPs is of ca. 4-5 nm. The clear lattice fringes and the selected area
electron diffraction patterns (SAED) (insets of Fig. 2a-c) indicate the good crystallinity of
6
ZnO QDs. This was further confirmed by the XRD patterns of ZnO QDs in which the
diffraction peaks could well be indexed to the wurtzite structure of ZnO (space group P63mc,
JCPDS No 36-1451) (Fig. 3). The broad peaks observed for all samples further confirm the
small size of ZnO NPs.
The UV-visible absorption spectra of ZnO QDs are shown in Fig. 2d-f and the excitonic
peaks are nearly located at the same wavelength (ca. 336 nm), further confirming that ZnO
QDs functionalized by APTMS, GLYMO or SBZ exhibit the same size. The excitonic
absorption of bulk ZnO appears at ca. 380 nm (3.26 eV), value significantly higher than those
determined for ZnO QDs, indicating the spatial confinement of the charge carriers in these
nanocrystals. Using the experimental formula [ref]:
1240/λ1/2 = a + b/D2 – c/D
where λ1/2 is the wavelength corresponding to an absorption of 50% to that of the excitonic
peak and a,b and c are parameters for ZnO QDs with diameters in the 2.5-6.5 nm range, the
diameters D of ZnO@APTMS, GLYMO and SBZ QDs were estimated to be xx, yy, and zz
nm, respectively, values in good agreement with those determined by TEM. Finally, due to
the intrinsic defects present in ZnO QDs prepared by a sol-gel method, the
photoluminescence (PL) spectra only exhibit deep-level visible emission centered at 557, 565
and 559 nm for of ZnO@APTMS, GLYMO and SBZ QDs, respectively (Fig. 2d-f). These
emissions originate from an electronic transition from a level close to the conduction band
edge to a defect associated trap state (oxygen vacancies, interstitial zinc sites or donor-
acceptor pairs).The PL quantum yields of ZnO@APTMS, GLYMO and SBZ QDs in water
were determined to be 21, 25 and 17%.
7
Fig. 2. TEM images of (a) ZnO@APTMS, (b) ZnO@GLYMO and (c) ZnO@SBZ QDs (the
insets are the SAED patterns). UV-visible absorption and PL emission spectra of (d)
ZnO@APTMS, (e) ZnO@GLYMO and (f) ZnO@SBZ QDs (the excitation wavelength is
330 nm).
Fig. 3. XRD patterns of ZnO QDs functionalized with APTMS, GLYMO or SBZ.
20 30 40 50 60 70 80
ZnO@APTMS
ZnO@GLYMO
Inte
ns
ity
(a
.u.)
2 (degree)
ZnO@SBZ
(10
0)
(00
2)
(10
1)
(10
2)
(11
0)
(10
3)
(112)
8
3.2. Surface chemistry-dependent ROS production of ZnO QDs
In this study, the toxicity of NPs-generated ROS was explored on the bacterial system, using
a set of ZnO core QDs. Four nanometer-sized ZnO core QDs were functionalized with either
3-aminopropyltrimethoxysilane (APTMS), (3-glycidyloxypropyl)trimethoxysilane
(GLYMO) or with a sulfobetaine zwitterionic (SBZ) ligand (Fig. 1) in order to engineer NPs
presenting different level of photo-induced ROS production, as well as different surface
chemistries enabling to develop different types of interactions with bacterial cells (Supporting
information, section 1). The three QDs exhibited different levels of photo-induced ROS
production (Fig. 1), therefore highlighting the importance of their surface chemistry on their
photoreactivity and possibly their toxicity if ROS production is to be concerned.
Fig. 4. Structure and photo-reactivity of the ZnO core QDs. ZnO@APTMS, ZnO@GLYMO,
and ZnO@SBZ QDs are all silane-based nanoparticles differing only by their surface
functionalization (amino groups, epoxides and a zwitterion surface coating respectively). The
ROS production following 1 h photoactivation was measured in water using common
procedures, and normalized to a same initial QD concentration for comparing ROS
production by the three nanoparticles (Supporting information, section 3).
3.2. Toxicity assessment of ZnO QDs on whole bacterial cells
The toxicity of the three ZnO QDs was evaluated on the E. coli strain MG1655(pUCD607)
using a bioluminescence extinction assay. A first set of experiments, carried out in the dark,
showed that the QDs exhibit similar toxicities, with Minimum Inhibitory Concentration
(MIC) values ranging from 1.8×10-4
to 2.9×10-4
MZnEq (molar concentration of zinc
equivalent). Noteworthy is that ZnO QDs are ca. twice less toxic than their equivalent content
in Zn2+
as seen with a ZnCl2 control (MIC ≈ 1×10-4
M; Fig. 5). With this respect, a dialysis
experiment carried out using ZnO@APTMS QDs at concentration slightly lower than MIC
ZnO QD structure Surface functionalization ROS production (µM)
!
[H2O2] [O2•-] [•OH]
ZnO@APTMS 19 161 980
ZnO@GLYMO 71 148 1490
ZnO@SBZ 70 330 220
9
(ca. 8×10-5
MZnEq) demonstrates that the concentration of free Zn2+
released does not
exceed the “No Observable Effect Concentration” (NOEC) obtained for ZnCl2 (ca. 1×10-5
M). This therefore rules out the possibility that the toxic effects observed were solely due to
the release of Zn2+
originating from the dissolution of QDs or Zn-containing adsorbed
complexes (Supporting information). In a second set of experiments, the QDs toxicity against
E. coli was investigated under simulated sunlight exposure. Despite a marked difference
regarding their photo-reactivity in terms of ROS production (Fig. 1), no significant difference
in toxicity on E. coli was observed between the three QDs, where MIC values were similar to
those observed in the dark (Fig. 2, Supporting information Fig. S2). Despite stable MIC
values, the bioluminescence profiles appeared to evolve with the illumination time. If part of
it could be attributed to the photo-sensitivity of bioluminescence production, as demonstrated
in Fig. 2D, the NOECs are clearly displaced towards lower concentrations in a QD-dependent
manner. It should be observed that according to Fig. 2C, the ZnO@GLYMO QDs, displaying
the lowest NOEC value after two hours of illumination, are also the higher hydroxyl radical
producers.
Fig. 2. Effect of light irradiation on ZnO core QD toxicity. Bioluminescence extinction
curves of strain E. coli MG1655(pUCD607) exposed to ZnCl2, ZnO@APTMS,
10-7 10-6 10-5 10-4 10-2 10-3
50
Dark conditions
10 min. illumination
120 min. illumination
10-7 10-6 10-5 10-4 10-2 10-3
1.105
2.105
3.105
4.105
5.105
6.105
7.105
0
Concentration (MZnEq)
0
25
75
100
150
125
Bio
lum
ines
ce
nc
e
(% o
f co
ntr
ol)
10-7 10-6 10-5 10-4 10-2 10-3
Concentration (MZnEq)
0
25
50
75
100
150
125
Bio
lum
ines
ce
nc
e
(% o
f co
ntr
ol)
A - Dark conditions B - Illumination - 10 min.
Concentration (MZnEq)
0
25
50
75
100
150
125
Bio
lum
ine
sc
en
ce
(%
of
co
ntr
ol)
Light conditions
Bio
lum
ine
sc
en
ce
(n
orm
alize
d t
o b
iom
ass
)
C - Illumination - 120 min. D - Light sensitivity of control
ZnCl2
ZnO@APTMS
ZnO@GLYMO
ZnO@SBZ
ZnCl2
ZnO@APTMS
ZnO@GLYMO
ZnO@SBZ
ZnCl2
ZnO@APTMS
ZnO@GLYMO
ZnO@SBZ
100%
143%
58%
10
ZnO@GLYMO, and ZnO@SBZ QDs maintained in the dark for 2 hours (A), maintained in
the dark for 110 min and then exposed to a 10 min illumination (B), or illuminated for 2 h
(C). Control assays without ZnCl2 or ZnO QDs amendment displayed a light sensitivity
depending of the exposure time (D). Further comparisons are available in the Supporting
information. Error bars represent minimum and maximum values of two independent sets of
experiments for (A), (B), (C), and standard deviation (n=8) for (D).
3.3. Stress assessment using whole cell biosensor assays
When considering the results of the bioluminescence-based toxicity tests, it clearly appears
that (i) the three ZnO QDs are less toxic than their Zn content, and (ii) the levels of Zn2+
leaking from the NPs are so low (Supporting information Fig. S1) that QDs dissolution
cannot be considered as a significant source of toxicity by itself. If Zn2+
leakage is of limited
incidence, it is tempting to propose ROS production as an alternative source of toxicity.
However, the stable MIC values observed regardless of the illumination conditions would
rather be in favor of a non-ROS-based toxicity. An elucidation of this dilemma was attempted
using a set of specific E. coli whole cell biosensors for Zn2+
, oxidative stress, and DNA
damages (Supporting information, section 6), in order to identify which cell stress dominates
when bacterial cells are exposed to ZnO QDs. The use of such biosensors non-ambiguously
demonstrated that only Zn2+
ion leakage was perceived by the bacteria (Supporting
information Fig. S3C and S3D), although Zn2+
concentration was too low to explain the QD
toxicity by itself. In addition, even if the three ZnO QDs produce ROS, no oxidative stress
response was observed using a dedicated biosensor (Supporting information Fig. S3A). All
together, the relative stability of the NPs and the absence of photo-induced toxicity tend to
show that neither Zn2+
leakage nor ROS production are responsible for the QDs toxicity. This
necessarily implies an alternative source of toxicity, which is not without reminding previous
suggestions regarding the deleterious effect of direct cell-NPs interactions [10,12,13]. Still,
the absence of ROS-induced stress response in the whole cell biosensors may be surprising as
ROS production could be measured for all QDs (Fig. 1). Since ROS are produced
extracellularly, cell impairments caused by the oxidative damages of bacterial surface
structures are likely to occur before cells could trigger any detectable stress response.
Alternatively, the non-ROS non-Zn2+
toxicity mentioned before may dominate the QDs
toxicity, which leads to strong cell damages that preclude subsequent oxidative stress
response to be triggered. At this stage, the level of damages generated by photo-induced ROS
remains an open question.
11
3.4. Setting up a DNA degradation assay to assess QD photo-induced reactivity
Apart from being limited to settings allowing physiological conditions, the use of live
material for toxicity testing requires that cells should be able to respond before being too
damaged. On the principle, this could be overcome if the analytical procedures focus on the
alteration of biological material by itself rather than the alteration of the cell functioning. In
the next series of experiments, DNA was used as a surrogate biomolecule to estimate the
QDs injuriousness at elevated concentrations (to prevent high dissolution rate), and in non-
physiological medium (to limit ROS scavenging if any). Five hundred ng fractions of H2O-
dissolved DNA (plasmid pBELX) were independently exposed for 10 min to each of the QDs
at concentrations ranging from 1.5 to 1.8 ×10-3 MZnEq, in the dark or under light irradiation.
An agarose gel electrophoresis of the mixtures clearly showed that in the dark, the QDs had
no effect on DNA integrity (Fig. 3A). Only in the case of the ZnO@APTMS QDs, both a
plasmid band reduction and a fluorescent smear above the plasmid band indicate a strong
interaction between the DNA and the QDs leading to a gel retardation. Conversely, a 10 min
light irradiation had a dramatic effect on DNA integrity where fluorescent smears below the
normal plasmid band size indicate random DNA fragmentation and thus severe DNA
damages. The levels of DNA alteration were further measured by quantitative PCR (qPCR),
considering that a full DNA integrity is a prerequisite for a proper PCR-amplification to
proceed. Analyses of DNA integrity by qPCR, after purification of the QDs-exposed DNA,
showed that, in the dark, ZnO@APTMS and ZnO@SBZ QDs poorly altered pBELX DNA
(ca. 75% integrity) while only 4% of undamaged DNA could be recovered after exposure to
ZnO@GLYMO. A 10 min light irradiation resulted in dramatic effects on DNA integrity,
with ca. 0.4-0.5% of unaltered pBELX recovered after exposure to ZnO@APTMS and
ZnO@SBZ QDs, and as low as 0.02% recovery in the case of the ZnO@GLYMO QDs.
These experiments clearly indicate that (i) despite being undetected in whole cell-based
toxicity tests, ZnO QDs could cause severe photo-induced damages on biomolecules, and (ii)
the levels of photo-induced damages observed partially follow the levels of ROS generated
by the various NPs, with ZnO@GLYMO being both the more reactive and the more
damaging QDs.
12
Fig. 3. Photo-induced DNA damages by ZnO QDs. a, Alteration of pBELX plasmid DNA22
(3169 bp) upon exposure to ZnO@APTMS (A), ZnO@SBZ (S), and ZnO@GLYMO (G) for
10 min with or without light irradiation (full spectrum white light, 9 klux illuminence) and
the mixture was run on a 1% agarose gel in TAE buffer. DNA was stained with ethidium
bromide and “2log DNA ladder” (New England Biolabs®) was used as DNA molecular
weight marker. An image analysis (Image Lab™ Software, Gel Doc™, Bio-Rad) shows that
the total amount of the fluorescence is conserved between lane except for from lanes #13,
#14, #15, where only 40% to 10% of the fluorescence is kept, indicating a pure loss of DNA
material. b, Quantification of pBELX DNA integrity following a 10 min exposure to
ZnO@APTMS (A), ZnO@SBZ (S), and ZnO@GLYMO (G) at 1.610-3, 1.510-3, and
1.810-3 MZnEq respectively, with or without light irradiation. DNA integrity was estimated
by qPCR and is given as a fraction (% recovery) of what obtained for untreated DNA
controls. Error bars represent standard deviations of 6 values (2 independent experiments
with triplicate qPCR assays each).
4. CONCLUSIONS
In conclusion, using a set of differently functionalized ZnO QDs, NPs generating various
levels of photo-induced ROS could be engineered. Although the highest ROS producers QDs
QDs
control
control before
incubation
10 min. in
the dark
10 min. light
irradiation
Em
pty
la
ne
DN
A la
dde
r
QDs A S G - A C G - A S G - A S G
DNA - - - + + + + + + + + + + + +
plasmid DNA
DNA
retardation
smear
DNA
degradation
smear
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17
Re
co
ve
ry o
f
un
alt
ere
d D
NA
(%
)
100
10
1
0.1
0.01
Un
tre
ate
d
DN
A
10 min. in the dark 10 min. light irradiation
QDs - A S G - A S G
DNA + + + + + + + +
A
B
13
(ZnO@GLYMO) were the most damaging towards DNA, the levels of photo-induced
damages did not fully correlate with the ROS production, indicating that alternative
phenomena such as QD-biomolecule interactions might also contribute to the overall NP
reactivity. However, if all ZnO QDs were undoubtedly very photo-reactive towards a
surrogate biomolecule (DNA), no clear photo-induced toxicity could be observed in whole
bacterial cells toxicity assays. This likely illustrates the barrier effect exerted by numerous
biomolecules present on cell surfaces that prevent the ROS produced to reach critical cell
functions. With this respect, it is worth mentioning that, contrary to eukaryotic cells, bacteria
do not practice endocytosis and QDs have to cross several external barriers before reaching
vital functions, suggesting that only closely delivered ROS could be effective. While the ROS
did not appear to be directly involved in the cell toxicity observed for the studied ZnO QDs,
it was quite surprising to observe that the free Zn2+
leaked by the QDs did not reach the
critical level where cell functioning could be impaired. Considering that both ROS
production and Zn2+
leaks are insufficient to impair cell functioning, a third source of toxicity
has to be considered. Taking into account the nanoscale dimensions of the particles, direct
and deleterious interactions of the NPs with important extracellular/periplasmic cell
components (e.g. respiratory chain, transporters) could well be a plausible explanation to
explore. Whatever the toxicity mechanisms involved, this study emphasizes the fact that
toxicity tests carried out on live bacterial cells are likely inappropriate for evaluating NP
generated-ROS deleterious effects. For that reason, the DNA-damaging test we propose here
advantageously offers (i) the possibility of quantifying photo-induced damages on a
biomolecule without making use of live cells therefore avoiding interfering parameters such
as NP internalization or bio-generated ROS, (ii) to work in non-physiological conditions, e.g.
in water, thus avoiding ROS scavenging for instance, and (iii) to work at NP concentrations
well over the toxic level if necessary.
ACKNOWLEDGMENT
This work was supported by the program CESA of the French National Research Agency
(ANR) (Project “NanoZnOTox”; ANR-11-CESA-0004ANR). Additional support was gained
from the Hydreos competitiveness cluster. The authors wish to thank Hervé Marmier (LIEC,
Université de Lorraine) for ICP-OES analyses, Ghouti Medjhadi (IJL, University de
Lorraine) for XRD analyses and Mathieu Lhuire for technical assistance.
14
APPENDIX A. SUPPLEMENTARY DATA
Supplementary data associated with this article can be found, in the online version.
REFERENCES
1. C.A. Charitidis, P. Georgiou, M.A. Koklioti, A.-F. Trompeta, V. Markakis. Manufacturing
nanomaterials: from research to industry. Manufacturing Rev. 1 (2014) 11.
2. W.J. Stark, P.R. Stoessel, W. Wohlleben, A. Hafnerc. Industrial applications of
nanoparticles. Chem. Soc. Rev. 44 (2015) 5793-5805.
3. Z.L. Wang. Zinc oxide nanostructures: growth, properties and applications. J. Phys.:
Condens. Matter 16 (2004) R829-R858.
4. H. Moussa, E. Girot, K. Mozet, H. Alem, G. Medjahdi, R. Schneider. ZnO rods/reduced
graphene oxide composites prepared via a solvothermal reaction for efficient sunlight-driven
photocatalysis. Appl. Catal. B: Environ. 185 (2016) 11-21.
5. X. Liu, J. Zhang, L. Wang, T. Yang, X. Guo, S. Wu, S. Wang. 3D hierarchically porous
ZnO structures and their functionalization by Au nanoparticles for gas sensors. J. Mater.
Chem. 21 (2011) 349-356.
6. M.M. Khin, A. Sreekumaran Nair, V. Jagadeesh Babu, R. Murugana, S. Ramakrishna. A
review on nanomaterials for environmental remediation. Energy Environ. Sci. 5 (2012) 8075-
8109.
7. Y. Li, W. Zhang, J. Niu, Y. Chen. Mechanism of photogenerated reactive oxygen species
and correlation with the antibacterial properties of engineered metal-oxide nanoparticles.
ACS Nano 6 (2012) 5164-5173.
8. H. Ma, L.K. Wallis, S. Diamond, S. Li, J. Canas-Carrell, A. Parra. Impact of solar UV
radiation on toxicity of ZnO nanoparticles through photocatalytic reactive oxygen species
(ROS) generation and photo-induced dissolution. Environ. Pollut. 2014, 193, 165-172.
9. V.L. Prasanna, R. Vijayaraghavan. Insight into the mechanism of antibacterial activity of
ZnO: Surface defects mediated reactive oxygen species even in the dark. Langmuir 31 (2015)
9155-9162.
15
10. A.B. Djurišić, Y.H. Leung, A.M. Ng, X.Y. Xu, P.K. Lee, N. Degger, R.S. Wu. Toxicity
of metal oxide nanoparticles: mechanisms, characterization, and avoiding experimental
artefacts. Small 11 (2015) 26-44.
11. J. Ying, T. Zhang, M. Tang Metal Oxide Nanomaterial QNAR Models: available
structural descriptors and understanding of toxicity mechanisms. Nanomaterials (Basel) 5
(2015) 1620–1637.
12. Y.H. Leung, A.M. Ng, X. Xu, Z. Shen, L.A. Gethings, M.T. Wong, C.M. Chan, M.Y.
Guo, Y.H. Ng, A.B. Djurišić, P.K. Lee, W.K. Chan, L.H. Yu, D.L. Phillips, A.P. Ma, F.C.
Leung. Mechanisms of antibacterial activity of MgO: non-ROS mediated toxicity of MgO
nanoparticles towards Escherichia coli. Small 10 (2014) 1171-1183.
13. K. Feris, C. Otto, J. Tinker, D. Wingett, A. Punnoose, A. Thurber, Kongara, M., M.
Sabetian, B. Quinn, C. Hanna, D. Pink. Electrostatic interactions affect nanoparticle-mediated
toxicity to gram-negative bacterium Pseudomonas aeruginosa PAO1. Langmuir, 26 (2010)
4429-4436.
14. X. Bellanger, P. Billard, R. Schneider, L. Balan, C. Merlin. Stability and toxicity of ZnO
quantum dots: interplay between nanoparticles and bacteria. J. Hazard. Mater. 283 (2015)
110-116.
15. J. Ying, T. Zhang, M. Tang. Metal Oxide Nanomaterial QNAR Models: available
structural descriptors and understanding of toxicity mechanisms. Nanomaterials (Basel) 5,
(2015) 1620–1637.
16. J.A. Lemire, J.J. Harrison, R.J. Turner. Antimicrobial activity of metals: mechanisms,
molecular targets and applications. Nat. Rev. Microbiol. 11 (2013) 371-384.
17. W. Song, J. Zhang, J. Guo, J. Zhang, F. Ding, L. Li, , Z. Sun. Role of the dissolved zinc
ion and reactive oxygen species in cytotoxicity of ZnO nanoparticles. Toxicol. Lett., 199
(2010) 389-397.
18. M. Li, L. Zhu, D. Lin. Toxicity of ZnO Nanoparticles to Escherichia coli: mechanism and
the influence of medium components. Environ. Sci. Technol. 45 (2011) 1977–1983.
19. G. Applerot, A. Lipovsky, R. Dror, N. Perkas, Y. Nitzan, R. Lubart, A. Gedanken,
Enhanced antibacterial activity of nanocrystalline ZnO due to increased ROS-mediated cell
injury. Adv. Funct. Mater. 19 (2009) 842–852.
16
20. D. Wang, L. Zhao, H.Ma, H. Zhang, L.-H. Guo. Quantitative analysis of reactive oxygen
species photogenerated on metal oxide nanoparticles and their bacteria toxicity: the role of
superoxide radicals. Environ. Sci. Technol. 51 (2017) 10137−10145.
21. X. Jiang, W. He, X. Zhang, Y. Wu, Q. Zhang, G. Cao, H. Zhang, J. Zheng, T.R. Croley,
J.-J. Yin. Light-induced assembly of metal nanoparticles on ZnO enhances the generation of
charge carriers, reactive oxygen species, and antibacterial activity. J. Phys. Chem. C 122
(2018) 29414−29425.
22. C. Mao, Y. Xiang, X. Liu, Z.Cui, X. Yang, K.W.K. Yeung, H. Pan, X. Wang, P.K. Chu,
S. Wu. Photo-inspired antibacterial activity and wound healing acceleration by hydrogel
embedded with Ag/Ag@AgCl/ZnO nanostructures. ACS Nano 11 (2017) 9010−9021.
23. H. Moussa, C. Merlin, C. Dezanet, L. Balan, G. Medjahdi, M. Ben-Attia, R. Schneider.
Trace amounts of Cu²⁺ ions influence ROS production and cytotoxicity of ZnO quantum dots.
J. Hazard. Mater. 304 (2016) 532-542.
24. R.K. Dutta, B.P.Nenavathu, M.K. Gangishetty, A.V.R. Reddy. Studies on antibacterial
activity of ZnO nanoparticles by ROS induced lipid peroxidation. Colloids Surf. B:
Biointerfaces 94 (2012) 143-150.
25. J. Ma, J.-Y. Chen, Y. Zhang, P.-N. Wang, J. Guo, W.-L. Yang, C.-C. Wang.
Photochemical instability of thiol-capped CdTe quantum dots in aqueous solution and living
cells: process and mechanism. J. Phys. Chem. B 111 (2007) 12012–12016.
26. F.A. Kauffer, C.Merlin, L. Balan, R. Schneider. Incidence of the core composition on the
stability, the ROS production and the toxicity of CdSe quantum dots, J. Hazard. Mater. 268
(2014) 246-255.
27. M. Arakha, M. Saleem, B.C. Mallick, S. Jha. The effects of interfacial potential on
antimicrobial propensity of ZnO nanoparticle. Sci. Rep. 5 (2015) 9578.
28. X. Bellanger, H. Guilloteau, B. Breuil, C. Merlin. Natural microbial communities
supporting the transfer of the IncP-1β plasmid pB10 exhibit a higher initial content of
plasmids from the same incompatibility group. Front. Microbiol. 5 (2014) 637.